Project Spotlight One: AI RAGKASTEN

By Chuck Hennet — September 10, 2025

What the heck is a RAGKASTEN, you ask?

Good question — because this one’s still taking shape.

The AI RAGKASTEN is a project in progress. It’s an experiment in building a personal research and synthesis tool — part Zettelkasten, part AI lab, part file cabinet duct-taped to a rocket.

Here’s the basic idea:

RAG stands for Retrieval-Augmented Generation — a method of helping large language models (LLMs) answer questions based on your own real-world data: notes, transcripts, spreadsheets, source code, documents. Instead of having the AI guess, you give it something to work with. The results are more grounded, more useful, and more accountable.

Kasten comes from Zettelkasten, a note-taking and knowledge-networking system that treats each note as a small, self-contained thought — heavily linked and built for emergence. It’s a tool not just for storage, but for thinking.

Put them together, and you get the RAGKASTEN:
A system where AI meets structure. A tool that helps you hold more ideas, link more fragments, and surface deeper connections.

This project is still unfolding. The file formats are evolving. The chunking strategy is in flux. The UX isn’t settled. But the goal is clear:
To build something that helps real humans synthesize messy knowledge — one note, one file, one connection at a time.

We’ll use this space to share that process as it happens: ideas, roadblocks, architecture, code, and the unexpected insights that come from trying to make something weird and useful.